Semantics Research Group Meeting, June 17, 2016

Focus-sensitive operators such as ‘only’ must associate with a focused constituent in their scope (Jackendoff, 1972; Rooth, 1985, 1992; a.o.). We present new arguments in support of the view that focus association involves covert focus movement with pied-piping (Drubig, 1994; Krifka, 2006; Wagner, 2006), and against the more prominent approach which does not involve movement (Rooth 1985, 1992).

The first argument comes from Tanglewood configurations of the form in Kratzer (1991). We show that Tanglewood configurations are island-sensitive, and Kratzer’s arguments to the contrary are overcome by the availability of covert pied-piping. Our results show that Tanglewood configurations must be the result of covert focus movement with pied-piping, and the focus-index mechanism proposed by Kratzer in fact is not an option in the grammar.

The second argument comes from the distribution of Beck (2006) focus intervention effects in focus association constructions. We show that certain quantifiers disrupt the interpretation of focus when above and near the in-situ focus. This pattern is predicted by the availability of covert focus movement with pied-piping, where focus intervention affects the interpretation of pied-piped constituents, but is unpredicted by the non-movement association approach. Time permitting, additional arguments from binding and parasitic gaps will be discussed.

(Joint work with Hadas Kotek.)